The Artificial University: Decision Support for Universities in the COVID-19 Era

Figure 2

Regression-subset diagram for TAU using the “minimize infections” metric on a university of 6,500 people. The single most important intervention is on the top row, explaining 70% of the variance in the outcome metric according to the adjusted-R² test. The second row shows the effect of adding the second most important explanatory factor into the regression (adjusted-R² lifts to 0.83). As each new factor is added, there are marginal gains in the adjusted-R² value.